1899.] PRINCE— PASSAMAQUODDY WITCHCRAFT TALES. 183 



become as brave as the fallen foe had been. All authorities tend to 

 show, however, that cannibalism was extremely rare amono- the 

 American races, and was only resorted to in isolated cases like the 

 one here noted. ^ 



The sixth tale, of the klwa'kzu, or snow-demon, is one of a great 

 number. The 'Algic Indian believed in many spirits, some benevo- 

 lent like the wnn,^''mesw^k or litde people, who were wont to warn 

 the tribesmen of impending danger,* some harmless like the wan- 

 dering klwa'kw or the chib'/akufe, the tree sprite, who sits in the 

 crotch of^the large branches, and some distinctly malevolent like 

 the appdd'mk\or spirit of the deep water who lurks in the lakes to 

 drag down the unwary swimmer. 



In the notation of the tales I have used the following system. 

 The consonants are to be pronounced as in English, with the 

 exception of / and k, which represent voiceless tennes ; n, which is 

 the French nasal n; w\ the whisded initial peculiar to the Pass., 

 Abn. and Lenape, and ', which Is a guttural voice-stop, not unlike the 

 Semitic ayin. When n and m are written in juxtaposition to a fol- 

 lowing consonant they have their simple nasal tone-value. The 

 vowels, whose exact quantity I have marked in the Passamaquoddy 

 and Penobscot, have the Italian values, except in the case of the 

 apostrophe. ', which is a very short u, and «, which is equivalent to oo 

 in "good." 



The intonation of tlrc Passamaquoddy dialect is difficult to 

 acquire. In the narrative style, the syllables are spoken in what is 

 nearly a monotone, until the tone syllable is reached, when the 

 voice runs up a musical third and drops the same interval 

 on the syllable after the accent. When a word has two accents, 

 one following the other, as, for example, in pdhegfmul, the first 

 accented "Syllable is unusually prolonged in a sing-song tone and 

 the second Is marked" by the voice-rise. These peculiarities are 

 perfectly reproduced by the phonograph. 



I have made the grammatical analysis following the tales, partly by 

 means of a direct study of the Pass, itself and pardy by means of a 

 •colloquial knowledge of the kindred Abenaki language of Canada, 

 which is almost identical with the Penobscot idiom of Oldtown, Me., 

 and is very close to the Lenape. 



*Cf. Prince, Annals N.Y. Acad. Sci., xi, p. 373. 



